Aggregate Morphology and Internal Structure for Asphalt Concrete: Prestep of Computer-Generated Microstructural Models

Abstract

The reconstruction of microstructural models is the prerequisite step for the micromechanical simulation of asphalt concrete. Computer-generated microstructural models have been developed because the traditional X-ray image-based models are costly and time-consuming. One concern for the computer-generated model is the accuracy in aggregate morphology and internal structure compared with the X-ray image-based models. Therefore, it is desirable to generate microstructural models with aggregate morphology and internal structures close to the real asphalt concrete samples. Although aggregate morphology and internal structures for real samples have been previously studied, these parameters for computer-generated models are rarely reported. This study aims to propose an approach to analyze and quantify the aggregate morphology and internal structures of computer-generated microstructural models of asphalt concrete. Sphere clumps were used to represent aggregate particles and air voids. Then the aggregate skeleton and air voids were generated based on which both discrete element (DE) models and finite-element models on the microscale can be reconstructed. The aggregate sphericity and angularity, air void distribution, aggregate orientation, aggregate contact points, and aggregate distribution were analyzed based on the DE model. Some novel approaches were proposed to quantify these parameters. The results showed that all these parameters can be successfully obtained and quantified. The outcomes of this study can serve as a prestep of the ultimate goal of reconstructing microstructural models that display aggregate morphology and an internal structure comparable to real asphalt concrete samples.